Amazon SageMaker Data Agent integrates business context into conversations

Amazon Web Services has enhanced its SageMaker Data Agent with new integration capabilities that allow the AI assistant to leverage business context and metadata from SageMaker Catalog when helping data practitioners discover datasets and generate code. The integration enables users to interact with data using natural business terminology rather than technical database names, with the agent able to search through glossary terms, custom metadata forms, asset summaries, and documentation to identify the correct tables and columns for analysis tasks. The enhanced Data Agent can now handle business-focused queries like "Calculate customer retention rate" or "What data do I have on customer churn?" and automatically generate more accurate SQL and Python code on the first attempt. The system leverages business context that organizations have curated in their SageMaker Catalog, including metadata synced from third-party data governance platforms like Collibra, Atlan, and Alation. Additionally, the agent respects data governance policies by checking subscription status and providing access request links when users lack proper permissions. The new functionality is available through SageMaker Unified Studio notebooks and Query Editor across all AWS regions where the service operates. AWS positions this enhancement as a way for organizations to maximize their existing catalog investments while reducing time-to-insight and enabling data teams to work more efficiently using business language rather than deciphering technical schema names.